An entity graph that round-trips data it does not understand, codecs that never import domain semantics, and geometry that lowers into a neutral kernel. No C++ in the dependency graph.
The model stores entities structurally rather than as domain structs. A file full of cost entities parses and re-exports intact in a build compiled with no cost crate at all — verified by tests, not asserted by prose.
How the separations work
A thin viewer takes default-features = false, features = ["step"] and compiles no domain code and no geometry stack. Domains and codecs are cargo features over one shared vocabulary.
Choosing features
Several crates are deliberate architecture scaffolds that own module names without implementing behaviour. The capability matrix distinguishes implemented, partial, scaffold, and absent — per entity.
Read the matrix
ifc-geometry answers what an IFC entity means geometrically and lowers implemented families into the format-neutral Axiolid DAG. It never triangulates, evaluates NURBS, or picks an execution provider.
The Axiolid boundary
[dependencies]
openbim-ifc = { git = "https://github.com/openbimrs/ifc.git", rev = "a7c4949bb941504ce874bdec13bd81d33491b5cb" }The workspace crates are not published on crates.io yet. Cargo locks this immutable Git source in Cargo.lock.
The library target is named ifc, so call sites read as a facade:
use ifc::{Codec, Model, StepCodec};
let model = StepCodec.read_bytes(source)?;
println!("{} entities", model.len());
# Ok::<(), ifc::ModelError>(())This project publishes a capability matrix that separates implemented behaviour from reserved module ownership. Some domain crates are currently scaffolds: files that own a name and a doc comment so the architecture is reviewable, without implementing the entity.
Read the matrix before scoping work. No capability should be inferred from a crate name, a module path, or an IFC entity appearing in the schema.
If you are evaluating this crate for a specific application, the use cases section works an end-to-end scenario against the real current state of the code, including what the application author must still build themselves.